FIFA 19 ANALYSIS¶
IMPORTING MODULES¶
In [1]:
### FOR MANUPULATION
import numpy as np
import pandas as pd
### FOR VISUALIZATION
import seaborn as sns
import matplotlib.pyplot as plt
import matplotlib.image as mpimg
import plotly as py
import plotly.graph_objs as go
import seaborn as sns
from IPython.display import display, HTML, YouTubeVideo
from collections import Counter as counter
print("Modules are imported")
FIFA19 COVER PHOTO¶
In [2]:
plt.figure(1, figsize = (20, 15))
img = mpimg.imread("fifa19cover.jpg")
plt.imshow(img)
plt.show()
LOADING DATASET¶
In [3]:
df_fifa = pd.read_csv("data.csv")
df_fifa.head(50)
Out[3]:
SHAPE OF DATASET(ROW, COLUMNS)
In [4]:
df_fifa.shape
Out[4]:
DROPPING UNNECESSARY COLUMNS
In [5]:
useless_cols = ["ID", "Unnamed: 0", "Photo", "Flag", "Club Logo", "Special", "Body Type", "Real Face", "Joined", "Loaned From",
"Contract Valid Until", "Weight", "LS", "ST", "RS", "LW", "LF", "CF", "RF", "RW", "LAM", "CAM", "RAM", "LM",
"LCM", "CM", "RCM", "RM", "LWB", "LDM", "CDM", "RDM", "RWB", "LB", "LCB", "CB", "RCB", "RB", "Release Clause"]
df_fifa.drop(useless_cols, axis=1, inplace=True)
df_fifa.head(10)
Out[5]:
INFORMATION ON DATASET
In [6]:
df_fifa.info()
STATISTICS ON DATASET
In [7]:
df_fifa.describe()
Out[7]:
ELEMENTS IN DATASETS
In [8]:
df_fifa.nunique()
Out[8]:
EXPLORATORY DATA ANALYSIS¶
HIGHEST RATED PLAYERS IN FIFA19
In [9]:
df_fifa.sort_values(by = "Overall", ascending = False)[["Name", "Position", "Nationality", "Club", "Age", "Overall"]].head()
Out[9]:
LOWEST RATED PLAYERS IN FIFA19
In [10]:
df_fifa.sort_values(by = "Overall", ascending = True)[["Name", "Position", "Nationality", "Club", "Age", "Overall"]].head()
Out[10]:
PLAYERS WITH BEST POTENTIAL
In [11]:
df_fifa.sort_values(by = "Potential", ascending = False)[["Name", "Position", "Potential", "Club", "Age", "Overall"]].head()
Out[11]:
OLDEST PLAYERS
In [12]:
df_fifa.sort_values(by = "Age", ascending = False)[["Name", "Position", "Potential", "Club", "Age", "Overall"]].head()
Out[12]:
YOUNGEST PLAYERS
In [13]:
df_fifa.sort_values(by = "Age", ascending = True)[["Name", "Position", "Potential", "Club", "Age", "Overall"]].head()
Out[13]:
PLAYERS WITH BEST SHOTPOWER
In [14]:
df_fifa.sort_values(by = "ShotPower", ascending = False)[["Name", "Club", "Position", "ShotPower", "Overall"]].head()
Out[14]:
BEST DRIBBLERS IN THE GAME
In [15]:
df_fifa.sort_values(by = "Dribbling", ascending = False)[["Name", "Club", "Position", "Dribbling", "Overall"]].head()
Out[15]:
BEST CROSSERS OF THE BALL
In [16]:
df_fifa.sort_values(by = "Crossing", ascending = False)[["Name", "Club", "Position", "Crossing", "Overall"]].head()
Out[16]:
BEST FREEKICK TAKERS
In [17]:
df_fifa.sort_values(by = "FKAccuracy", ascending = False)[["Name", "Club", "Position", "FKAccuracy", "Overall"]].head()
Out[17]:
PLAYERS HAVING BEST BALL CONTROL
In [18]:
df_fifa.sort_values(by = "BallControl", ascending = False)[["Name", "Club", "Position", "BallControl", "Overall"]].head()
Out[18]:
PLAYERS WITH BEST VISION
In [19]:
df_fifa.sort_values(by = "Vision" , ascending = False)[["Name", "Club" ,"Nationality" ,"Vision", "Overall" ]].head()
Out[19]:
BEST FINISHERS
In [20]:
df_fifa.sort_values(by = "Finishing", ascending = False)[["Name", "Club", "Position", "Finishing", "Overall"]].head()
Out[20]:
PLAYERS WITH FAST ACCELERATION
In [21]:
df_fifa.sort_values(by = "Acceleration", ascending = False)[["Name", "Club", "Position", "Acceleration", "Overall"]].head()
Out[21]:
BEST SPRINTERS
In [22]:
df_fifa.sort_values(by = "SprintSpeed", ascending = False)[["Name", "Club", "Position", "SprintSpeed", "Overall"]].head()
Out[22]:
BEST LONG PASSERS IN THE GAME
In [23]:
df_fifa.sort_values(by = "LongPassing", ascending = False)[["Name", "Club", "Position", "LongPassing", "Overall"]].head()
Out[23]:
STRONGEST PLAYERS
In [24]:
df_fifa.sort_values(by = "Strength", ascending = False)[["Name", "Club", "Position", "Strength", "Overall"]].head()
Out[24]:
MOST AGGRESIVE PLAYERS
In [25]:
df_fifa.sort_values(by = "Aggression", ascending = False)[["Name", "Club", "Position", "Aggression", "Overall"]].head()
Out[25]:
BEST GOAL KEEPERS (REFLEXES)
In [26]:
df_fifa.sort_values(by = "GKReflexes", ascending = False)[["Name", "Club", "Position", "GKReflexes", "Overall"]].head()
Out[26]:
BEST GOAL KEEPERS (DIVING)
In [27]:
df_fifa.sort_values(by = "GKDiving", ascending = False)[["Name", "Club", "Position", "GKDiving", "Overall"]].head()
Out[27]:
BEST DEFENDERS (SLIDING TACKLE)
In [28]:
df_fifa.sort_values(by = "SlidingTackle", ascending = False)[["Name", "Club", "Nationality", "SlidingTackle", "Overall"]].head()
Out[28]:
BEST DEFENDERS (STANDING TACKLE)
In [29]:
df_fifa.sort_values(by = "StandingTackle", ascending = False)[["Name", "Club", "Nationality", "StandingTackle", "Overall"]].head()
Out[29]:
NUMBER OF PLAYERS AT CERTAIN POSITIONS
In [30]:
plt.figure(1, figsize = (15, 6))
pos = sns.countplot(x = "Position", data = df_fifa, palette = "hls");
pos.set_title(label="Number of players at specialized position", fontsize=40);
plt.show()
BEST LEFT FOOTED PLAYERS
In [31]:
df_fifa[df_fifa["Preferred Foot"] == "Left"][["Name", "Club", "Position","Overall"]].head()
Out[31]:
BEST RIGHT FOOTED PLAYERS
In [32]:
df_fifa[df_fifa["Preferred Foot"] == "Right"][["Name","Club", "Position", "Overall"]].head()
Out[32]:
COMPARISON : LEFT FOOT VS RIGHT FOOT IN TERMS OF DRIBBLING AND FINISHING
In [33]:
sns.lmplot(x = "Finishing", y = "Dribbling", data = df_fifa, scatter_kws = {"alpha":0.1}, col = "Preferred Foot");
plt.show()
ATTRIBUTES : PLAYER NAME
In [34]:
cols = ["Crossing", "Potential", "Finishing", "HeadingAccuracy", "ShortPassing", "Volleys",
"Dribbling", "Curve", "FKAccuracy", "LongPassing", "BallControl", "Acceleration",
"SprintSpeed", "Agility", "Reactions", "Balance", "ShotPower", "Jumping", "Stamina",
"Strength", "LongShots", "Aggression", "Interceptions", "Positioning", "Vision",
"Penalties", "Composure", "Marking", "StandingTackle", "SlidingTackle", "GKDiving",
"GKHandling", "GKKicking", "GKPositioning", "GKReflexes"]
i=0
while i < len(cols):
print('Best {0} : {1}'.format(cols[i], df_fifa.loc[df_fifa[cols[i]].idxmax()]["Name"]))
i += 1
BEST PLAYER AT EACH POSITION
In [35]:
display(HTML(df_fifa.iloc[df_fifa.groupby(df_fifa["Position"])["Overall"].idxmax()][["Name", "Position"]].to_html(index=False)))
TOP 3 FEATURES FOR EACH POSITION
In [36]:
player_features = ["Crossing", "Potential", "Finishing", "HeadingAccuracy", "ShortPassing", "Volleys",
"Dribbling", "Curve", "FKAccuracy", "LongPassing", "BallControl", "Acceleration",
"SprintSpeed", "Agility", "Reactions", "Balance", "ShotPower", "Jumping", "Stamina",
"Strength", "LongShots", "Aggression", "Interceptions", "Positioning", "Vision",
"Penalties", "Composure", "Marking", "StandingTackle", "SlidingTackle", "GKDiving",
"GKHandling", "GKKicking", "GKPositioning", "GKReflexes"]
for i, val in df_fifa.groupby(df_fifa["Position"])[player_features].mean().iterrows():
print("Position {}: {}, {}, {}".format(i, *tuple(val.nlargest(3).index)))
FREQUENCY OF OVERALL PLAYER RATINGS
In [37]:
plt.figure(1, figsize = (15, 6))
pos = sns.countplot(x = "Overall", data = df_fifa, palette = "hls");
pos.set_title(label="Player Ratings out of 100", fontsize=40);
plt.show()
PLAYER POTENTIAL DISTRIBUTION
In [38]:
plt.figure(1, figsize = (15, 6))
pos = sns.countplot(x = "Potential", data = df_fifa, palette = "hls");
pos.set_title(label="Player Potential Distribution", fontsize=40);
plt.show()
PLAYERS AGE DISTRIBUTION
In [39]:
plt.figure(1, figsize = (15, 6))
pos = sns.countplot(x = "Age", data = df_fifa, palette = "hls");
pos.set_title(label="Age Distribution", fontsize=40);
plt.show()
AGE DISTRIBUTION IN TOP CLUBS IN EUROPE
In [40]:
clubs = ["FC Barcelona", "Real Madrid", "Juventus", "Liverpool", "Manchester City", "Paris Saint-Germain"]
df_club_age = df_fifa.loc[df_fifa["Club"].isin(clubs) & df_fifa["Age"]]
plt.figure(1 , figsize = (15 ,7))
sns.violinplot(x = "Club", y = "Age", data = df_club_age)
plt.title("Age Distribution in Top European clubs")
plt.xticks(rotation = 50)
plt.show()
CLUBS WITH MOST PLAYERS
In [41]:
print("Total number of clubs : {0}".format(df_fifa["Club"].nunique()))
print(df_fifa["Club"].value_counts().head(10))
BEST SQUAD SIZE
In [42]:
plt.figure(1 , figsize = (15 , 5))
club = []
c = counter(df_fifa["Club"]).most_common()[:11]
for n in range(11):club.append(c[n][0])
sns.countplot(x = "Club", data = df_fifa[df_fifa["Club"].isin(club)],
order = df_fifa[df_fifa["Club"].isin(club)]["Club"].value_counts().index, palette = "rocket")
plt.xticks(rotation = 90)
plt.title("Clubs with best squad size" )
plt.show()
CHANGING DATA FORMAT
In [43]:
def cleaning_value(x):
if '€' in str(x) and 'M' in str(x):
c = str(x).replace('€' , '')
c = str(c).replace('M' , '')
c = float(c) * 1000000
else:
c = str(x).replace('€' , '')
c = str(c).replace('K' , '')
c = float(c) * 1000
return c
fn = lambda x : cleaning_value(x)
df_fifa["Value_num"] = df_fifa["Value"].apply(fn)
df_fifa["Wage_num"] = df_fifa["Wage"].apply(fn)
MOST VALUABLE PLAYERS
In [44]:
df_fifa.sort_values(by = "Value_num", ascending = False)[["Name", "Club", "Nationality", "Overall", "Value"]].head()
Out[44]:
BEST PAID PLAYERS
In [45]:
df_fifa.sort_values(by = "Wage_num", ascending = False)[["Name", "Club", "Nationality", "Overall", "Wage"]].head()
Out[45]:
AGE DISTRIBUTION IN TOP COUNTRIES
In [46]:
country = ["Argentina", "Brazil", "France", "Belgium", "England", "Spain", "Portugal"]
df_country_age = df_fifa.loc[df_fifa["Nationality"].isin(country) & df_fifa["Age"]]
plt.figure(1 , figsize = (15 ,7))
sns.violinplot(x = "Nationality" , y = "Age" , data = df_country_age)
plt.title("Age Distribution in Top National Teams")
plt.xticks(rotation = 50)
plt.show()
AGE VS OVERALL RATING COMPARISON
In [47]:
plt.figure(1 , figsize = (15 , 6))
sns.boxplot(df_fifa["Age"], df_fifa["Overall"])
plt.title("Age vs Overall rating Comparison")
plt.show()
AGE VS STAMINA COMPARISON
In [48]:
plt.figure(1 , figsize = (15 , 9))
sns.boxplot(x = "Age", y = "Stamina", data = df_fifa)
plt.title("Age vs Stamina Comparison")
plt.show()
AGE VS SPRINT SPEED COMPARISON
In [49]:
plt.figure(1 , figsize = (15 , 9))
sns.boxplot(x = "Age", y = "SprintSpeed", data = df_fifa)
plt.title("Age vs Sprint Speed Comparison")
plt.show()
AGE VS REACTION COMPARISON
In [50]:
plt.figure(1 , figsize = (15 , 9))
sns.boxplot(x = "Age", y = "Reactions", data = df_fifa)
plt.title("Age vs Reactions Comparison")
plt.show()
DRIBBLING VS FINISHING COMPARISON
In [51]:
plt.figure(1 , figsize = (25 , 9))
sns.boxplot(x = "Dribbling", y = "Finishing", data = df_fifa)
plt.title("Dribbling vs Finishing Comparison")
plt.show()
AGE VS FINISHING COMPARISON
In [52]:
plt.figure(1 , figsize = (15 , 9))
sns.boxplot(x = "Age", y = "Finishing", data = df_fifa)
plt.title("Age vs Finishing Comparison")
plt.show()
COMPARING FEATURES WITH RESPECT TO ACCELERATION
In [53]:
def make_scatter(df_fifa):
feats = ("Agility", "Balance", "Dribbling", "SprintSpeed")
for index, feat in enumerate(feats):
plt.subplot(len(feats)/4+1, 4, index+1)
ax = sns.regplot(x = 'Acceleration', y = feat, data = df_fifa)
plt.figure(figsize = (20, 20))
plt.subplots_adjust(hspace = 0.4)
make_scatter(df_fifa)
plt.show()
COUNTRY WITH MOST PLAYERS
In [54]:
print("Total number of countries : {0}".format(df_fifa["Nationality"].nunique()))
print(df_fifa["Nationality"].value_counts().head(10))
COUNTPLOT DISTRIBUTION
In [55]:
plt.figure(1, figsize = (20 , 7))
countries = []
c = counter(df_fifa["Nationality"]).most_common()[:11]
for n in range(11) : countries.append(c[n][0])
sns.countplot(x = "Nationality", data = df_fifa[df_fifa["Nationality"].isin(countries)],
order = df_fifa[df_fifa["Nationality"].isin(countries)]["Nationality"].value_counts().index,
palette = "rocket")
plt.xticks(rotation = 90)
plt.title("Most Players Belong To" )
plt.show()
PIE CHART DISTRIBUTION
In [56]:
countries=df_fifa["Nationality"].value_counts()
index=countries.index
con=pd.DataFrame({"Country":index,"Count":countries})
con=con["England":"Netherlands"]
plt.pie(con["Count"],labels=con["Country"],wedgeprops = {"linewidth": 5},autopct='%1.2f%%')
plt.title("Distribution of players in top 10 countries")
plt.tight_layout()
plt.show()
COMPARISON - FC BARCELONA, REAL MADRID, LIVERPOOL AND MANCHESTER CITY¶
FC BARCELONA¶
In [57]:
plt.figure(1, figsize = (15, 10))
img = mpimg.imread("Barcelonalogo.jpg")
plt.imshow(img)
plt.show()
FULL SQUAD LIST OF FC BARCELONA
In [58]:
df_fifa[df_fifa["Club"] == "FC Barcelona"][["Name", "Position", "Overall", "Age", "Wage", "Nationality"]]
Out[58]:
THE BEST PLAYER OF FC BARCELONA¶
LIONEL MESSI HAS BEEN RANKED AT THE TOP OF FIFA19 RATINGS
HE IS ALSO THE BEST PLAYER OF FC BARCELONA
In [59]:
plt.figure(1, figsize = (15, 10))
img = mpimg.imread("Messi.jpg")
plt.imshow(img)
plt.show()
REAL MADRID¶
In [60]:
plt.figure(1, figsize = (15, 20))
img = mpimg.imread("rmlogo.png")
plt.imshow(img)
plt.show()
FULL SQUAD LIST OF REAL MADRID
In [61]:
df_fifa[df_fifa["Club"] == "Real Madrid"][["Name", "Position", "Overall", "Age", "Wage", "Nationality"]]
Out[61]:
BEST PLAYER OF REAL MADRID¶
LUKA MODRIC IS THE HIGHEST RATED REAL MADRID PLAYER IN FIFA19
In [62]:
plt.figure(1, figsize = (15, 20))
img = mpimg.imread("modric.jpg")
plt.imshow(img)
plt.show()
LIVERPOOL FOOTBALL CLUB¶
In [63]:
plt.figure(1, figsize = (15, 20))
img = mpimg.imread("liverpoollogo.jpg")
plt.imshow(img)
plt.show()
FULL SQUAD LIST OF LIVERPOOL
In [64]:
df_fifa[df_fifa["Club"] == "Liverpool"][["Name", "Position", "Overall", "Age", "Wage", "Nationality"]]
Out[64]:
BEST PLAYER OF LIVERPOOL¶
M. SALAH IS HIGHEST RATED LIVERPOOL PLAYER IN FIFA19
In [65]:
plt.figure(1, figsize = (15, 20))
img = mpimg.imread("salah.jpg")
plt.imshow(img)
plt.show()
MANCHESTER CITY¶
In [66]:
plt.figure(1, figsize = (15, 20))
img = mpimg.imread("mancitylogo.jpg")
plt.imshow(img)
plt.show()
FULL SQUAD LIST OF MANCHESTER CITY
In [67]:
df_fifa[df_fifa["Club"] == "Manchester City"][["Name", "Position", "Overall", "Age", "Wage", "Nationality"]]
Out[67]:
BEST PLAYER OF MANCHESTER CITY¶
KEVIN DE BRUYNE IS THE HIGHEST RATED MANCHESTER CITY PLAYER IN FIFA19
In [68]:
plt.figure(1, figsize = (15, 20))
img = mpimg.imread("kevin.jpg")
plt.imshow(img)
plt.show()
DEFINING NEW VARIABLE NAME FOR THE CLUBS
In [69]:
barca = df_fifa[df_fifa["Club"] == "FC Barcelona"]
madrid = df_fifa[df_fifa["Club"] == "Real Madrid"]
liverpool = df_fifa[df_fifa["Club"] == "Liverpool"]
mancity = df_fifa[df_fifa["Club"] == "Manchester City"]
NUMBER OF FC BARCELONA PLAYERS AVAILABLE FOR EACH POSITION
In [70]:
barca["Position"].value_counts()
Out[70]:
NUMBER OF REAL MADRID PLAYERS AVAILABLE FOR EACH POSITION
In [71]:
madrid["Position"].value_counts()
Out[71]:
NUMBER OF LIVERPOOL PLAYERS AVAILABLE FOR EACH POSITION
In [72]:
liverpool["Position"].value_counts()
Out[72]:
NUMBER OF MANCHESTER CITY PLAYERS AVAILABLE FOR EACH POSITION
In [73]:
mancity["Position"].value_counts()
Out[73]:
DISTPLOT : AGE OF PLAYERS
In [74]:
plt.figure(1, figsize = (15, 6))
sns.distplot(barca["Age"], color = "blue")
sns.distplot(madrid["Age"], color = "black")
sns.distplot(liverpool["Age"], color = "red")
sns.distplot(mancity["Age"], color = "white")
plt.title("Comparison of distribution of Age between Top Club players")
plt.tight_layout()
plt.show()
DISTPLOT : PLAYER RATING
In [75]:
plt.figure(1, figsize = (15, 6))
sns.distplot(barca["Overall"], color = "blue")
sns.distplot(madrid["Overall"], color = "black")
sns.distplot(liverpool["Overall"], color = "red")
sns.distplot(mancity["Overall"], color = "white")
plt.title("Comparison of distribution of Player Ratings between Top Club players")
plt.tight_layout()
plt.show()
COMPARING PLAYER VALUE DISTRIBUTION AMONG CLUBS
In [76]:
plt.figure(1, figsize = (15, 6))
sns.distplot(barca["Value_num"], color = "blue")
sns.distplot(madrid["Value_num"], color = "black")
sns.distplot(liverpool["Value_num"], color = "red")
sns.distplot(mancity["Value_num"], color = "white")
plt.title("Comparison of Player Value between in Top Clubs")
plt.tight_layout()
plt.show()
COMPARING PLAYER WAGE DISTRIBUTION
In [77]:
plt.figure(1, figsize = (15, 6))
sns.distplot(barca["Wage_num"], color = "blue")
sns.distplot(madrid["Wage_num"], color = "black")
sns.distplot(liverpool["Wage_num"], color = "red")
sns.distplot(mancity["Wage_num"], color = "white")
plt.title("Comparison of Player Wages between Top Club")
plt.tight_layout()
plt.show()
COMPARING PLAYER POTENTIAL
In [78]:
plt.figure(1, figsize = (15, 6))
sns.distplot(barca["Potential"], color = "blue")
sns.distplot(madrid["Potential"], color = "black")
sns.distplot(liverpool["Potential"], color = "red")
sns.distplot(mancity["Potential"], color = "white")
plt.title("Distribution of Player Potential between Top Club")
plt.tight_layout()
plt.show()
COMPARING PLAYERS FINISHING
BETTER FINISHING LEADS TO MORE GOALS SCORED
In [79]:
plt.figure(1, figsize = (15, 6))
sns.distplot(barca["Finishing"], color = "blue")
sns.distplot(madrid["Finishing"], color = "black")
sns.distplot(liverpool["Finishing"], color = "red")
sns.distplot(mancity["Finishing"], color = "white")
plt.title("Distribution of Players Finishing capabilities between Top Club")
plt.tight_layout()
plt.show()
COUNTPLOT : PLAYER DISTRIBUTION WITH RESPECT TO COUNTRY
In [80]:
plt.figure(1, figsize = (15, 6))
plt.subplot(1, 4, 1)
sns.countplot(barca["Nationality"], color = "blue")
plt.xticks(rotation=90)
plt.title("FC Barcelona")
plt.subplot(1, 4, 2)
sns.countplot(madrid["Nationality"], color = "black")
plt.xticks(rotation=90)
plt.title("Real Madrid")
plt.subplot(1, 4, 3)
sns.countplot(liverpool["Nationality"], color = "red")
plt.xticks(rotation=90)
plt.title("Liverpool")
plt.subplot(1, 4, 4)
sns.countplot(mancity["Nationality"], color = "green")
plt.xticks(rotation=90)
plt.title("Manchester City")
plt.tight_layout()
plt.show()
In [81]:
cols = ["Preferred Foot", "International Reputation", 'Skill Moves']
for col in cols:
plt.subplot(1, 4, 1)
sns.countplot(barca[col])
plt.title("FC Barcelona")
plt.subplot(1, 4, 2)
sns.countplot(madrid[col])
plt.title("Real Madrid")
plt.subplot(1, 4, 3)
sns.countplot(liverpool[col])
plt.title("Liverpool")
plt.subplot(1, 4, 4)
sns.countplot(mancity[col])
plt.title("Manchester City")
plt.tight_layout()
plt.show()
BRAIN BEHIND SUCCESSFULL TEAMS¶
THE COACH/MANAGERS¶
LEFT : PEP GUARDIOLA, CLUBS : FC BARCELONA, FC BAYERN, MANCHESTER CITY
CENTER : ZINEDINE ZIDANE, CLUB : REAL MADRID
RIGHT : JURGEN KLOPP, CLUBS : BORUSSIA DORTMUND, LIVERPOOL
In [91]:
plt.figure(1, figsize = (15, 20))
img = mpimg.imread("managers.jpg")
plt.imshow(img)
plt.show()
THEY PLAY A VITAL PART IN TEAMS SUCCESS